How to remove the empty result set caused by bucket_selector?

Elastic Stack Elasticsearch
1

This operation will show you my problem.

1、Create Index

PUT car
{
  "mappings": {
    "properties": {
      "color": {
        "type": "keyword"
      },
      "company": {
        "type": "keyword"
      },
      "price":{
        "type": "double"
      }
    }
  }
}

2、Import data

POST car/_doc
{
  "color":"red",
  "company":"byd",
  "price":120000
}

POST car/_doc
{
  "color":"blue",
  "company":"byd",
  "price":130000
}

POST car/_doc
{
  "color":"blue",
  "company":"bmw",
  "price":230000
}

3、Do Search

GET car/_search
{
  "size": 0,
  "aggs": {
    "company": {
      "terms": {
        "field": "company"
      },
      "aggs": {
        "color": {
          "terms": {
            "field": "color"
          },
          "aggs": {
            "price": {
              "avg": {
                "field": "price"
              }
            },
            "price_bucket_filter": {
              "bucket_selector": {
                "buckets_path": {
                  "price": "price"
                },
                "script": "params.price > 200000"
              }
            }
          }
        }
      }
    }
  }
}

The search query will return

{
  "took" : 189,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 3,
      "relation" : "eq"
    },
    "max_score" : null,
    "hits" : [ ]
  },
  "aggregations" : {
    "company" : {
      "doc_count_error_upper_bound" : 0,
      "sum_other_doc_count" : 0,
      "buckets" : [
        {
          "key" : "byd",
          "doc_count" : 2,
          "color" : {
            "doc_count_error_upper_bound" : 0,
            "sum_other_doc_count" : 0,
            "buckets" : [ ]  # how to remove the empty result?
          }
        },
        {
          "key" : "bmw",
          "doc_count" : 1,
          "color" : {
            "doc_count_error_upper_bound" : 0,
            "sum_other_doc_count" : 0,
            "buckets" : [
              {
                "key" : "blue",
                "doc_count" : 1,
                "price" : {
                  "value" : 230000.0
                }
              }
            ]
          }
        }
      ]
    }
  }
}

What should I do if I wish to get the company and color of a car that sells for $200,000 or more? Now the "byd" in this query is invalid and I don't want him to appear.

I tried to add a second 'bucket_selector', but he doesn't filter the array length.

GET car/_search
{
  "size": 0,
  "aggs": {
    "company": {
      "terms": {
        "field": "company"
      },
      "aggs": {
        "color": {
          "terms": {
            "field": "color"
          },
          "aggs": {
            "price": {
              "avg": {
                "field": "price"
              }
            },
            "price_bucket_filter": {
              "bucket_selector": {
                "buckets_path": {
                  "price": "price"
                },
                "script": "params.price > 200000"
              }
            }
          }
        },
        "empty_bucket_filter": {
          "bucket_selector": {
            "buckets_path": {
              "len": "color.buckets.length"
            },
            "script": "params.len>0"
          }
        }
      }
    }
  }
}
2

What about this?

GET car/_search
{
  "size": 0,
  "query": {
    "range": {
      "price": {
        "gte": 200000
      }
    }
  }, 
  "aggs": {
    "company": {
      "terms": {
        "field": "company"
      },
      "aggs": {
        "color": {
          "terms": {
            "field": "color"
          }
        }
      }
    }
  }
}

It gives:

{
  "took": 3,
  "timed_out": false,
  "_shards": {
    "total": 1,
    "successful": 1,
    "skipped": 0,
    "failed": 0
  },
  "hits": {
    "total": {
      "value": 1,
      "relation": "eq"
    },
    "max_score": null,
    "hits": []
  },
  "aggregations": {
    "company": {
      "doc_count_error_upper_bound": 0,
      "sum_other_doc_count": 0,
      "buckets": [
        {
          "key": "bmw",
          "doc_count": 1,
          "color": {
            "doc_count_error_upper_bound": 0,
            "sum_other_doc_count": 0,
            "buckets": [
              {
                "key": "blue",
                "doc_count": 1
              }
            ]
          }
        }
      ]
    }
  }
}
3

Sorry, maybe I didn't express myself clearly. What I expect is to filter by the average price of the colors.

4

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.